Discovery of a Novel Series of Biphenyl Benzoic Acid Derivatives as

4002

J. Med. Chem. 2008, 51, 4002–4020

Discovery of a Novel Series of Biphenyl Benzoic Acid Derivatives as Highly Potent and Selective Human β3 Adrenergic Receptor Agonists with Good Oral Bioavailability. Part II Masashi Imanishi,† Shinji Itou,† Kenichi Washizuka,† Hitoshi Hamashima,† Yutaka Nakajima,† Takanobu Araki,† Yasuyo Tomishima, Minoru Sakurai,† Shigeo Matsui,‡ Emiko Imamura,‡ Koji Ueshima,§ Takao Yamamoto,‡ Nobuhiro Yamamoto,‡ Hirofumi Ishikawa, Keiko Nakano,‡ Naoko Unami,‡ Kaori Hamada,‡ Yasuhiro Matsumura,| Fujiko Takamura,| and Kouji Hattori*,† Chemistry Research Laboratories, Pharmacological Research Laboratories, Applied Pharmacology Research Laboratories, and Analysis & Pharmacokinetic Research Laboratories, Astellas Pharma Inc., 21 Miyukigaoka, Tsukuba-shi, Ibaraki 305-8585, Japan ReceiVed January 15, 2008

The left-hand side (LHS) and central part of our first generation biphenyl (FGB) series was modified to improve in vitro and in vivo β3-AR potency without loss of oral bioavailability. First, in this study, we focused our efforts on replacement of the 3-chlorophenyl moiety in the LHS of FGB analogues with 3-pyridyl ring analogues to adjust the lipophilicity. Second, we investigated the replacement of the central part of this series and discovered that introduction of a methyl group into the R-position of the phenethylamine moiety greatly enhanced potency keeping good oral availability. Finally, the replacement of the two carbon linker of the central part with an ethoxy-based linker provided improved potency and PK profiles. As a result of these studies, several analogues (i.e., 9h, 9k, 9l, 10g, 10m, 10p, 10r, 11b, and 11l) were identified that displayed an excellent balance of very higher human β3-AR potency compared to the FGB compounds, high selectivity, and good pharmacokinetic profiles. Furthermore, these several compounds showed high in vivo efficacy in an overactive bladder (OAB) model. These findings suggest that our selected second generation biphenyl (SGB) series compounds may be attractive new successful therapeutic candidates for the treatment of OAB. Introduction The β3-adrenergic receptor (β3-AR)a has been shown to mediate various pharmacological and physiological effects such as lipolysis, thermogenesis,1 intestinal smooth muscle relaxation,2 and urinary bladder detrusor muscle relaxation.3,4 Thus, the activation of the human β3-AR has attracted much attention as a potential approach toward the treatment of obesity, noninsulin-dependent diabetes mellitus (NIDDM), irritable bowel syndrome, and overactive bladder, and the β3-AR is therefore recognized as an attractive target for drug discovery.5 Recently, on the other hand, β3-AR selectivity over β1-AR and β2-AR is also important because stimulation of β1-AR and β2AR may induce severe side effects such as enhancement of heart rate and tracheal relaxation, respectively. In the past decade, drug discovery efforts have shifted toward the design of selective agonists for the β3-AR. Furthermore, several groups have reported a number of potent and selective human β3-AR chemotypes (see Figure 1), but these are still not sufficient in terms of the pharmacokinetic properties.5b,6 Previous work in our laboratory has described the discovery of a series of first generation biphenyl (FGB) analogues containing a benzoic acid moiety on the right-hand side (RHS), represented by 8 (Figure 2), which exhibited good oral bio* To whom correspondence should be addressed. Phone: 81-29-863-7179. Fax: 81-29-852-5387. E-mail: [email protected]. † Chemistry Research Laboratories. ‡ Pharmacological Research Laboratories. § Applied Pharmacology Research Laboratories. | Analysis & Pharmacokinetic Research Laboratories. a Abbreviations: β-AR, β-adrenergic receptors; OAB, overactive bladder; FGB, first generation biphenyl; SGB, second generation biphenyl; LHS, left-hand side; RHS, right-hand side; cAMP, cyclic adenosine monophosphate; ISP, isoproterenol; CHO, Chinese hamster ovary; IVP, intravesical pressure; PAMPA, parallel artificial membrane permeation assay; PB, protein binding.

availability and a long plasma half-life.7 The structure-activity relationship (SAR) studies at the R position of the terminal phenyl ring on the RHS indicated that introduction of more lipophilic substitution increased β3-AR activity (O-cyclohexyl, 8c > O-iso-pr, 8b > O-Me, 8a) but decreased oral bioavailability (8a > 8b > 8c) (see Figure 2). On the basis of these results, we selected lead candidate 8b with a good balance of potency, selectivity, and pharmacokinetic profile. We extended optimization of the FGB series to further improve β3-AR activity without loss of the good oral bioavailability. Our designed second generation biphenyl (SGB) series (9-11) is shown in Figure 3. First of all, we focused our efforts on replacement of the 3-chlorophenyl moiety in the left-hand side (LHS) with several identified partial structures such as present in compounds 1,8 2,9 3,10 and 411 (see Figure 1). Second, we investigated the replacement of the central part of this series, through introduction of a methyl groups adjacent to the secondary amino group, such as in compounds 512 and 6.13 Finally, we modified the two carbon linker region, such as in compounds 4 and 714 (see Figure 1). In this paper, we describe synthesis and SAR studies in which we have varied the LHS and central part of the SGB series and simultaneously evaluated the pharmacokinetic profile by cassette dosing assay in dogs, as previously described. These studies have led to the successful discovery of several clinical drug candidates with an excellent balance of very high potency, selectivity and good pharmacokinetic profile. Chemistry. As shown in the first and second reactions of Scheme 1, in general, the requisite left and center part intermediate Boc amine derivatives (17-19) were synthesized by coupling of amino ethanol derivatives (12,15 13, 14) with carboxylic acid derivatives (15, 16) to afford the corresponding amide intermediate, followed by selective reduction of the amide moiety with BH3 · SMe2 to unmask the amino ethanol, followed

10.1021/jm8000345 CCC: $40.75  2008 American Chemical Society Published on Web 06/14/2008

Biphenyl Benzoic Acid DeriVatiVes. Part II

Journal of Medicinal Chemistry, 2008, Vol. 51, No. 13 4003

Figure 1. Structures of some β-3 AR agonists

Figure 2. First generation biphenyl (FGB) series.

by protection of the amine with a Boc group. In a similar way, the requisite intermediate 4-iodophenyl derivatives (24-26) were synthesized by coupling of commercially available mandelic acid derivatives 20 or 21 with amines 22, 23, which were synthesized as outlined in Scheme 2. The requisite intermediate Boc amine derivatives 28-32 containing a phenyl ethanol moiety were prepared as shown in the third and fourth reactions of Scheme 1. Coupling of optically active epoxides,16,17 in the presence of BSU with 4-bromophenyl ethylamine 27, followed by protection of the amine with a Boc

group, gave the corresponding Boc amine derivatives 28-32. In a similar manner, the requisite intermediate 4-iodophenyl derivatives (38-41) were synthesized by reaction of optically active epoxides 33 or 34, which are commercially available, or 2-chloro-5-[(2R)-2-oxiranyl]pyridine 35 prepared through known synthetic procedures, and with the chiral amines 36 or 37, and subsequently protection with a Boc group, respectively. As shown in Scheme 2, the optically active amine intermediate 36 containing a chiral methyl group was prepared in five steps starting from commercially available (2S)-2-amino-3-phenyl-

4004 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 13

Imanishi et al.

Figure 3. Second generation biphenyl (SGB) series.

1-propanol 47. The optically active amino propanol intermediate 37 was prepared by reduction of commercially available (2S)2-amino-3-(4-iodophenyl)propanoic acid with NaBH4 in the presence of H2SO4. As shown in the fifth and sixth reactions of Scheme 1, the intermediate Boc amine derivatives 43 and 44 containing an amino-pyridine moiety in the left part were prepared. Coupling of tosylate 42, which was prepared through known synthetic procedures,18 with 4-bromophenyl ethylamine 27 or 4-iodophenoxyethylamine 22 afforded a secondary amine, which was protected by protected with Boc to give the corresponding Boc amine derivatives 43 and 44. 4-Hydroxyphenyl intermediate 46 was prepared by coupling of commercially available (aS,bR)4-hydroxynorephedrine 45 with 4-bromophenylethylbromide, followed by protection with a Boc group. The general synthetic route to biphenyl targets (9a,b,h,j,k, 10a-i,g-k, 11a-k) is shown in Scheme 3. Suzuki crosscoupling of Boc amine intermediates with boronic acids (53a-h), the syntheses of which have been previously described,7 followed by alkaline hydrolysis of the methyl ester, and deprotection of the Boc group with 4 N HCl provided the target compounds as hydrochloride salts. In a similar manner, pyridine analogues (10p-s) were obtained from 40, in an additional step, through dechlorination by catalytic hydrogenation in the presence of HCO2NH4. The amino pyridine analogues (9i,l,m, 11l) were prepared by coupling of 43 or 44 with boronic acids (53b,e,f) in the presence of a catalytic amount of PdCl2(dppf) · CHCl3, followed by alkaline hydrolysis, and subsequent deprotection of the Boc amine silyl ether using 4 N HCl. The preparation methods for the final targets 9c-f are shown in Scheme 4. The phenol analogue 9c was obtained through a Suzuki coupling of Boc amine derivative 30 and boronic acid 53b, followed by deprotection of the benzyl group by catalytic hydrogenation, followed by using the same methods as described for 9a. In a similar manner, the aniline analogues 9e,f were obtained from 31 and 32, through an additional step of reduction of the nitro group with iron powder in the presence of NH4Cl. The methane sulfonamide analogue 9d was obtained from 31, in an additional step, through acylation of the NH2 group with Ms-Cl. The preparation of the final target 9g is shown in Scheme 5. The requisite intermediate 56 containing the methane sulfonamide moiety was obtained by reduction of the corresponding

nitro derivative 55, followed by coupling with Ms-Cl. The nitro derivative 55 was prepared from chiral oxirane 54, which has been previously described19 similar to procedures described for the third and fourth reactions of Scheme 1 using Cbz-Cl instead of Boc2O. The target 9g was obtained through a Suzuki coupling of Cbz derivative 56 and boronic ester 58, which was prepared from bromide 57 followed by deprotection of the Cbz group by catalytic hydrogenation. As shown in Scheme 6, the target 10j with a dimethyl group was synthesized from 61, similar to the preparation of 10d. The requisite intermediate 61 was prepared by protection of the amino group of commercially available 59 with trifluoro-acetyl, followed by para-selective iodination. Results and Discussion All compounds were evaluated for the ability to produce cAMP in Chinese hamster ovary (CHO) cell lines expressing cloned human β3 and β1-ARs. Selected compounds were also evaluated for human β2 activity using a similar method, as previously described.7,20 The results for reference compound, isoproterenol (ISP, nonselective β-AR agonist) are shown for comparison in Table 1. In addition, pharmacokinetic properties of selected compounds were evaluated by cassette dosing assay in dogs.7,21 As shown in Figure 2 and Table 1, in a previous article, we reported that our leading candidate 8b showed potent β3-AR activity (EC50 ) 1.1 nM), good selectivity relative to β1 and β2 activity, and good pharmacokinetics in all three species examined (rat, dog, and monkey). The O-cyclohexyl analogue (8c) with enhanced lipophilicity, resulted in a further improvement in β3-AR potency (EC50 ) 0.46 nM) but poor bioavailability. We also investigated the removal of the chloro atom on the LHS phenyl ring. More lipophilic isobutyl (8e) and O-chex (8f) analogues, relative to the O-iso-pr derivative (8d), showed improved β3-AR activity and selectivity compared to 8d. However, in the cassette dosing assay, both analogues displayed decreased Cmax levels. (These compounds showed high passive permeability in PAMPA.) These results indicated SAR trends as for the 3-chlorophenyl analogues that we have previously reported.7 First of all, to improve the β3-AR activity of 8b, we focused on replacement of the LHS 3-chlorophenyl moiety with several substituted phenyl groups. Shift of the chloro group to the



Scheme 1. Preparation of Left and Center Part Intermediatesa

a For first and second reactions: (a) HOBt, WSCD, DMF; (b) BH3 · SMe2, THF, DMI, then c. HCl; (c) (Boc)2O, THF, aq NaOH (pH ) 7-8). For third and fourth reactions: (a) BSU, DMSO, 65 °C, then (Boc)2O, THF, H2O; (b) EtOH, reflux; (c) (Boc)2O, THF, H2O, 1 N NaOH aq. For fifth and sixth reactions: (a) i-Pr2NEt, DMSO or DMF, 80 °C; (b) (Boc)2O, THF.

4-position (9a) and the fluoro analogue (9b) resulted in slightly decreased β3-AR activity relative to 8b. We investigated replacement of the chloro group with a hydroxy group at the 3-position, such as in compounds 1 or 6 (Figure 1). As a result, the phenol analogue (9c) resulted in 18-fold increased potency

(EC50 ) 0.062 nM) for β3 relative to 8b and high selectivity for β1. However, 9c showed lower Cmax levels relative to 8b in the cassette dosing assay. On the basis of this result, our efforts were focused on improving the PK properties of 9c by replacing the phenol part with isosteric functionalities. The methylsul-


Imanishi et al.

Scheme 2. Preparation of Alkyl Amine of Center Part Intermediatesa

(a) PPh3, 40% DEAD in toluene, 4 °C-room temp, then 4 N HCl/AcOEt; (b) HOBt, WSCD, DMF, then 4 N HCl/AcOEt; (c) CF3CO2Et, MeOH; (d) MsCl, Et3N, THF; (e) Zn(powder), NaI, AcOH, DME, reflux; (f) I2, HIO4 · 2H2O, AcOH, H2SO4, H2O, 80 °C; (g) 1 N NaOH aq, dioxane, 50 °C; (h) NaBH4, THF, H2SO4, Et2O. a

Scheme 3. General Synthesis Route to Targets 9, 10, 11a

a (a) Pd(PPh3)4, aq NaHCO3, DME, 70 °C; (b) 1 N NaOH aq, EtOH, THF, then 4 N HCl/AcOEt; (c) PdCl2(dppf) CH2Cl2, dppf, aq Na2CO3, toluene, EtOH, 75 °C (d) 1 N NaOH aq, EtOH, THF, then HCO2NH4, 10% Pd/C, MeOH, H2O, reflux, then 4 N HCl/dioxane; (e) PdCl2(dppf) CH2Cl2, dppf, aq Na2CO3, DMF, 80 °C; (f) 1 N NaOH aq, EtOH, 100 °C, then 4 N HCl/dioxane, MeOH.



Scheme 4. Synthesis Route to Targets 9c-fa

a (a) Pd(PPh3)4, aq NaHCO3, DME, 70 °C; (b) H2 (gas), 10% Pd/C, MeOH; (c) 1 N NaOH aq, EtOH, THF, then 4 N HCl/AcOEt (d) Fe (powder), NH4Cl, EtOH, reflux, (e) MsCl, pyridine.

Scheme 5. Synthesis Route to Targets 9ga

a (a) BSU, DMSO, 65 °C, then Cbz-Cl, THF, H2O; (b) Fe (powder), NH4Cl, EtOH, reflux, then MsCl, pyridine;(c) KOAc, pinacol diborane, PdCl2 (dppf)-CHCl3, dioxane, 90 °C; (d) Pd(PPh3)4, aq NaHCO3, DME, 70 °C; (e) H2 (gas), 10% Pd/C, MeOH, then 4 N HCl/AcOEt.

Scheme 6. Synthesis Route to Targets 10ja

a (a) (CF3CO)2O, Et3N, THF; (b) I2, HIO4 · 2H2O, AcOH, H2SO4, H2O, 80 °C; (c) 1 N NaOH aq, dioxane, 50 °C, then 34, EtOH, reflux; (d) (Boc)2O, THF, H2O, 1 N NaOHaq (pH ) 8-8.5); (e) 53b, Pd(PPh3)4, aq NaHCO3, DME, 70 °C; (f) 1 N NaOH aq, EtOH, THF, then 4 N HCl/AcOEt.

fonamide analogue 9d resulted in a substantial loss of potency (EC50 ) 30 nM) for β3 relative to 9c, and the Cmax level was not improved. The 3-aniline analogue 9e showed significantly decreased potency for β3 (EC50 ) 8.1 nM). The 4-aniline analogue 9f resulted in slightly decreased β3-AR activity (EC50 ) 2.6 nM) relative to 8b, but 9f had an improved Cmax level relative to the phenol analogue 9c. 4-Hydroxy-3-methylsulfonamide analogue 9g displayed stronger β3-AR activity but lower Cmax levels relative to 8b. We next investigated the replacement of LHS the 3-chlorophenyl ring with pyridine derivatives. The pyridine analogue 9h resulted in similar β3-AR activity (EC50 ) 1.5 nM) relative to 8b and 8d. Next, an amino-pyridine derivative was prepared and examined. As a result, compound 9i had greatly increased β3-AR activity (EC50 ) 0.19 nM) relative to the corresponding pyridine analogue 9h and phenyl analogues 8b and 8d, respectively. In addition, the pyridine analogue 9h and the amino-pyridine analogue 9i showed acceptable Cmax levels relative to 8b and 8d in the cassette dosing assay. On the basis of these findings, we attempted further optimization of the R2 substituent of pyridine and amino-pyridine analogues. Both the pyridine analogue 9h and the amino-

pyridine analogue 9i had reduced lipophilicity (9h, C log P ) 1.09; 9i, C log P ) 0.79) relative to chloro-phenyl 8b and phenyl 8d analogues (8b, C log P ) 3.30; 8d, Clog P ) 2.58). By analogy to the phenyl ring analogues (8e,f), the 3-pyridine derivatives with R2 ) isobutyl (9j) and O-c-hex (9k) were prepared, respectively. As predicted, the isobutyl analogue 9j (EC50 ) 0.26 nM) and the O-c-hex analogue 9k (EC50 ) 0.26 nM) exhibited higher potent β3-AR activity and selectivity relative to the O-iso-pr analogue 9h. Furthermore, compounds 9j and 9k were evaluated in the cassette dosing assay. It is noteworthy that these more lipophilic compounds (9j,9k) displayed acceptable Cmax levels relative to the O-iso-pr analogue 9h. In particular, the O-c-hex analogue (9k) showed a remarkable improvement of Cmax level relative to the same O-c-hex group analogues 8c and 8f. In consideration of the superior PK profile of 9k (9k, C log P ) 2.28) compared with 8c or 8f (8c, C log P ) 4.49; 8f, C log P ) 3.73), adjusting the lipophilicity by incorporation of the pyridine ring to the LHS may result in the improved PK profile. Therefore, the pyridine analogue 9k provided the best combination of high potency and Cmax level.


Imanishi et al.

Table 1. Effect of Conversion of Left-Hand Side of SGB Analogues

R1

compd

X

R2

human β3 EC50, nM a (IAb)

human β1 EC50, nM a

β1/β3

human β2 EC50, nM a

β2/β3

cassette (po) c Cmax ratio d

C log Pe

8b 8c 8d 8e 8f

3-Cl 3-Cl H H H

CH CH CH CH CH

O-iso-Pr O-c-Hex O-iso-Pr iso-Bu O-c-Hex

1.1 ( 0.1 (0.98) 0.46 ( 0.1 (1.0) 2.0 ( 0.06 (0.97) 0.60 ( 0.12 (0.99) 0.30 ( 0.02 (1.0)

720 ( 106 55 ( 5 >1000 >1000 260 ( 45

654 120 >500 >1667 867

>10000 NT >10000 >10000 NT

>9090 NT >5000 >16670 NT

1.0 0.0 0.90 0.25 0.0

3.30 4.49 2.58 3.09 3.73

9a 9b 9c 9d 9e 9f 9g 9h 9i 9j 9k 9l 9m

4-Cl 3-F 3-OH 3-NHMs 3-NH2 4-NH2 3-NHMs 4-OH H 4-NH2 H H 4-NH2 4-NH2

CH CH CH CH CH CH CH N N N N N N

O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr iso-Bu O-c-Hex iso-Bu O-c-Hex

2.4 ( 0.03 (0.96) 2.1 ( 0.2 (0.99) 0.062 ( 0.04 (0.99) 30 (0.74) 8.1 ( 0.6 (0.99) 2.6 ( 0.3 (0.98) 0.26 ( 0.03 (1.0) 1.5 ( 0.1 (0.97) 0.19 ( 0.02 (1.0) 0.26 ( 0.01 (0.99) 0.26 ( 0.02 (1.0) 0.066 ( 0.004 (0.97) 0.035 ( 0.005 (0.99)

500 ( 40 320 ( 38 440 ( 26 >1000 >1000 >1000 150 ( 8.8 >1000 130 ( 11 >1000 480 ( 52 150 ( 5 39 ( 0.5

208 152 7100 >33 >123 >380 577 >667 760 >3846 1769 2300 1114

NT NT NT NT NT NT NT >10000 >10000 NT >10000 3200 1100

NT NT NT NT NT NT NT >6667 >58800 NT >38400 48480 31420

NT 0.61 0.10 0.15 NT 0.35 0.06 0.65 0.49 0.36 0.48 0.36 0.09

3.30 2.73 1.92 1.39 1.36 1.36 0.73 1.09 0.76 1.71 2.28 1.27 1.95

0.97 ( 0.14 (1.0)

0.084 ( 0.02

0.087

2.0 ( 0.9

2.1

NT

ISP

f

The results are shown as the mean ( SE (n ) 3) or presented as the average of two experiments. b The intrinsic activity (IA) was defined as the ratio between the maximal effect of test compound and the maximal effect produced by isoproterenol. c Dose 0.1 mg/kg po (n ) 2-3). See References section for further details. d The ratio was defined between the Cmax of test compounds and the Cmax of 1b. The ratio value of 1b was presented as 1.0. e Biobyte C log P version 4.3. f Isoproterenol. NT: not tested. a

On the other hand, the amino-pyridine derivatives with R2 ) isobutyl (9l) and O-c-hex (9m) had greatly increased β3-AR activity (9l, EC50 ) 0.066 nM; 9m, EC50 ) 0.035 nM) relative to the parent pyridine analogues (9j,9k), respectively. In addition, the isobutyl analogue 9l maintained Cmax levels relative to the pyridine analogues and the corresponding O-iso-pr analogue 9i. Compound 9l showed the best profile of potency, selectivity, and C max level in Table 1. Unfortunately, the most potent amino-pyridine analogue with a O-c-hex moiety, 9m, showed poor Cmax levels in spite of the moderate C log P level (9m, C log P ) 1.95). Second, as can be seen in Figure 3, we shifted our attention to modification of biphenyl series 10, in which substituents were introduced at the R-position of the secondary amino group, such as in compounds 5 and 6 (see Figure 1), to improve β3-AR activity relative to 8b. From studies on the β3-AR activity of the four optical isomers of 512 and {2-[4[(2R)-2-[[(2R)-2-(3-chlorophenyl)-2-hydroxyethyl]amino]propyl]phenoxy]}acetic acid (BRL-37344),22 the (R,R)-configuration was shown to be important for enhancing β3-AR activity. As can be seen in Table 2, introduction of a methyl group into the R-position of the phenethylamine moiety of our nonsubstituted biphenyl series gave the (R)-methyl optical isomer, which exhibited more potent β3-AR activity (10a, EC50 ) 0.18 nM) compared with the corresponding (S) -methyl isomer (10b, EC50 ) 6.0 nM). By analogy to previous SAR studies, introducing at the R2 position of the terminal phenyl ring on the RHS with an O-iso-pr group maintained β3-AR activity (10c, EC50 ) 0.19 nM) and improved β3/β1 selectivity (10c, β3/β1 ) 195) relative to 10a. The nonsubstituted LHS phenyl ring derivatives with R2 ) H (10d), OMe (10e), OEt (10f), O-i-pr (10g), O-i-butyl (10h), and O-c-hex (10i) were next prepared and examined. Similar to the previous SAR trends, the order of potency at the R2 position

was OMe (10e) < H (10d), OEt (10f) < O-i-pr (10g) < O-ibutyl (10h), O-c-hex (10i) and selectivity for β1/β3 was H (10d), OMe (10e) < OEt (10f) < O-i-pr (10g), O-c-hex (10i) < O-i-butyl (10h). On the other hand, in the cassette dosing assay, the order of their Cmax levels was OMe (10e), OEt (10f) > O-i-pr (10g) . O-i-butyl (10h), O-c-hex (10i). These results also showed the same trends as previously demonstrated. Furthermore, these compounds (10f, 10g, 10h, 10i) were evaluated in the cassette dosing assay (iv) in dogs, and their pharmacokinetic parameters are shown in Table 3. The results indicated that introduction of more lipophilic substitution at the R2 position increased total clearance (10f, OEt < 10g, O-i-pr , 10h, O-i-butyl < 10i, O-c-hex) and decreased oral exposure (AUC) and bioavailability (10f, OEt > 10g, O-i-pr . 10h, O-i-butyl, 10i, O-c-hex). Actually, both O-ipr analogue 10g and O-c-hex analogue 10i are predicted to have good passive permeability as previously reported (based on PAMPA data) and, in liver microsomes, 10g and 10i showed good stability to dog and other species in terms of in vitro clearance (see Table 4). These results suggested that the high total clearance of the O-i-butyl (10h) and O-c-hex (10i) analogues may be due to a conjugation metabolism and/ or elimination, hence these compounds showed poor Cmax levels and oral exposure. On the other hand, compound 10g displayed an excellent balance of high β3-AR potency, high selectivity, and good pharmacokinetic profiles. Furthermore, replacement of the methyl group of 10g with a dimethyl group (10j) resulted in a 10-fold decrease in β3-AR activity. Compound 10k, having a hydroxy methyl group (Rconfiguration)23 exhibited significantly decreased potency but improved Cmax levels in the cassette dosing assay relative to 10g. The Kissei group have studied a series of 4′-hydroxynorepherine derivatives, such as compound 6 (see Figure 1). In our study, compound 10l having a 4′-hydroxynorepherine



Table 2. Effect of Conversion of the Central Part of SGB Analogues

R1

compd

X

A

8b

3-Cl

CH H

H

O-iso-Pr 1.1 ( 0.1 (0.98)

10a 10b 10c 10d 10e 10f 10g 10h 10i 10j 10k 10l 10m 10n 10o 10p 10q 10r 10s

3-Cl 3-Cl 3-Cl H H H H H H H H 4-OH H H H H H H H

CH CH CH CH CH CH CH CH CH CH CH CH CH CH CH N N N N

Me (R) Me (S) Me (R) Me (R) Me (R) Me (R) Me (R) Me (R) Me (R) Me2 CH2OH(R) H H H H Me (R) Me (R) Me (R) Me (R)

H H O-iso-Pr H O-Me O-Et O-iso-Pr O-iso-Bu O-c-Hex O-iso-Pr O-iso-Pr O-iso-Pr O-iso-Pr O-n-Pr iso-Bu H O-iso-Pr O-n-Pr O-c-Hex

H H H H H H H H H H H Me (S) Me (S) Me (S) Me (S) H H H H

human β1 EC50, human β2 EC50, nM a (IAb) β1/β3 nM a

human β3 EC50, nM a (IAb)

R2

B

0.18 ( 0.02 (0.96) 6.0 ( 0.4 (0.98) 0.19 ( 0.02 (0.96) 0.30( 0.02 (0.93) 0.79 ( 0.02 (0.90) 0.31 ( 0.07 (0.97) 0.091 ( 0.01 (1.01) 0.041 ( 0.001 (0.97) 0.042 ( 0.006 (1.07) 0.91 ( 0.09 (1.0) 7.4 ( 1 (0.98) 0.14 ( 0.01 (0.97) 0.81 ( 0.04 (0.97) 0.43 ( 0.07 (0.98) 0.14 ( 0.01 (0.98) 0.49 ( 0.005 (0.98) 0.099 ( 0.002 (1.0) 0.064 ( 0.002 (1.0) 0.069 ( 0.01 (1.06)

β2/β3

cassette (po) c Cmax ratio d

720 ( 106

654

>10000

>9090

1.0

13 ( 2 12 ( 1 37 ( 2 22 ( 2 83 62 86 ( 7 130 ( 17 28 ( 7 280 ( 38 >1000 500 ( 9.8 780 ( 26 690 ( 75 300 ( 17 75 ( 5 80 ( 10 190 ( 15 24 ( 4

72 2 195 73 105 200 945 3170 667 307 >135 3570 980 1607 2124 153 808 2930 348

NT NT NT >1000 NT NT >10000 NT NT NT NT NT >10000 >10000 NT >1000 >1000 >1000 NT

NT NT NT >3300 NT NT >10000 NT NT NT NT NT >12300 >23200 NT >2040 >10000 >15625 NT

NT NT 0.80 NT 2.82 2.83 0.63 0.01 0.00 0.43 1.37 0.04 0.42 0.13 0.16 2.3 0.58 0.29 NT

a The results are shown as the mean ( SE (n ) 3). b The intrinsic activity (IA) was defined as the ratio between the maximal effect of test compound and the maximal effect produced by isoproterenol. c Dose 0.1 or 0.2 mg/kg po (n ) 2-3). See References section for further details. d The ratio was defined between the Cmax of test compounds and the Cmax of 8b. The ratio value of 8b was presented as 1.0. NT: not tested.

Table 3. Pharmacokinetic Profiles of Compounds 10f-i in Dogs

a,b

po, (n ) 2-3) compd 10f 10g 10h 10i

iv, (n ) 2-3)

R

AUC0-2h (ng · h/mL)

T1/2β (hr)

Vdss(L/kg)

AUC0-24h (ng · h/mL)

CLtot (mL/min/kg)

F (%)c

O-Et O-iso-Pr O-iso-Bu O-c-Hex

350 ( 60 106 0.2 ( 0.2 0

9.9 15.5 17.7 ( 0.9 0.35

1.9 5.4 30.8 ( 3.4 1.13

477 260 44.5 ( 3.0 24.3

3.5 6.5 39.0 ( 2.4 68.9

73 41 0.4 0

a Cassette assay data. The results are shown as the mean ( SE (n ) 3) or presented as the average of two experiments. b Dose 0.1 mg/kg. po and iv (n ) 2-3). c F ) bioavailability.

Table 4. In Vitro Metabolism in Liver Microsomes CLint (mL/min/kg)a,b compd 9l 10g 10i

rat

dog

monkey

human

Discovery of a Novel Series of Biphenyl Benzoic Acid Derivatives as

Recommend Documents